Skip to content

Context Hub

A CLI that gives coding agents curated, versioned API docs and local notes they can reuse.

Stars
14k
Forks
1.2k
Licence
MIT
Language
JavaScript
Last update
5 months ago
Type
CLI

What it is

Context Hub is a command-line tool meant to be used by a coding agent rather than by a person directly. It lets the agent search and fetch maintained, versioned API documentation in markdown, which the README says helps reduce invented APIs. The agent can also attach local notes to docs and send up or down votes that go to doc authors.

  • Have an agent fetch the Python or JavaScript variant of OpenAI chat docs with chub get before writing code
  • Search for a payments or other API's docs with chub search and then fetch the relevant entry
  • Attach a local annotation such as a webhook workaround so a later session can see it with --with-annotations
  • Vote a doc up or down with chub feedback to send a rating to its maintainers

Great for

  • Developers who use a coding agent and want it to read current API docs before writing code
  • Teams that want to inspect and version-control the exact documentation their agent reads
  • Developers who want to contribute markdown docs or feedback back to a shared library
  • Users of Claude Code or similar agents that support skills via SKILL.md

Not ideal for

  • People looking for a graphical app to browse docs, since the README says Chub is designed for agents rather than direct human use
  • Projects that need docs for APIs not covered by the repo's content, since coverage depends on contributions
  • Anyone who cannot run Node.js 18 or newer, since the README lists that requirement
Step-by-step guide: how to use Context HubInstall, first run, examples, tips and FAQs

How to get it

  1. 1

    Install with a package manager

    You'll need: Node.js 18 or newer

    Install the CLI globally with npm

    npm install -g @aisuite/chub
  2. 2

    Run it

    • Install the package with npm install -g @aisuite/chub
    • Run chub search openai to list matching docs
    • Run chub get openai/chat --lang py to fetch the Python docs
    • Point your coding agent at the chub command or the SKILL.md file so it uses the tool
  3. 3

    Get the most out of it

    • Use --file or --full with chub get to fetch only the reference files you need
    • Use chub annotate to record gaps or workarounds, and pass --with-annotations on later fetches
    • Send chub feedback up or down so doc authors can improve entries
    • Install the get-api-docs SKILL.md so your agent is prompted to use Chub consistently
    • Read docs/cli-reference.md for the full list of commands, flags, and piping patterns

Downloads

Strengths

  • Docs are plain markdown in the repo, so you can read exactly what the agent sees
  • Supports language-specific variants such as --lang py and --lang js
  • Fetching can be limited to specific reference files with --file, or pulled in full with --full

Trade-offs

  • Annotations are treated as untrusted input by default, so agents must handle them carefully
  • The README says other content types such as agent skills are only on the roadmap, so the current scope is narrower
  • The description and usage rely on a coding agent being set up to call the CLI, which takes configuration

Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.